Review of Image Expression of Content-based Image Retrieval Technology

Abstract

With the continuous development of science and technology, the popularity and use of projectors, cameras and other electronic products have led to an ever-expanding image source. Due to the enormous workload of text-based image retrieval (TBIR) annotation, content-based image retrieval (CBIR) was proposed to solve this problem. The CBIR has experienced the process featured by an integration of image digital, image semantic, and other multiple image features. CBIR, however, still has shortcomings in embodying the high-level semantics of the image. As the ability of autonomously learning image features, the defects of CBIR in image expression can be effectively solved by deep learning, which thus has become a hot topic in current image expression researches. On the basis of content-based image retrieval technology, some readings and analyses are made in the article for exploring the future development of image retrieval technology in image expression field.